Moto g24 bootloader unlock ideas #1
Replies: 12 comments 94 replies
-
I think for this @R0rt1z2 can help us better finding an exploit. I verified that g24 lk doesn't require any key to unlock, but it checks for Secure Boot it seems. This is something that has to be studied with mtkclient, logs and decompilation, hoping that is possible to flash seccfg with mtkclient.
|
Beta Was this translation helpful? Give feedback.
-
Empty and writable partitions of G24. (Didn't test writing the userdata.)
|
Beta Was this translation helpful? Give feedback.
-
@cxzstuff Can you send the flash tool that downloads RSA? |
Beta Was this translation helpful? Give feedback.
-
"mrdump" Maybe this might come handy, from the name I'd say that it dumps memory. If it turns out right, it will surely be useful if we need to craft an exploit. I assume there's a fastboot command associated with it. |
Beta Was this translation helpful? Give feedback.
-
DA SLA keys for g24 got merged into mtkclient, now everyone can use it |
Beta Was this translation helpful? Give feedback.
-
We need to try to make the other DA file bypass SLA:
I've found it in the flash tool. So the dl_forbidden has the seccfg blocked, and it worked beforehand. |
Beta Was this translation helpful? Give feedback.
-
@cxzstuff can you try setting the "EnableLog" to true, and then use the console mode of the spft, and finally see if you get a log file here: |
Beta Was this translation helpful? Give feedback.
-
@progzone122 I want to try something with g24: if we dump the seccfg partition with mtkclient, we can try then flashing it with fastboot. I'm not sure if it'll work (probably not), but if it's the same file (thus same signature), what's stopping it from being flashed? Before asking the others to try this I want your opinion though. I know that messing with seccfg can result in a brick (I don't know if it's hardbrick or softbrick), so surely we might need to first understand how to parse it correctly. |
Beta Was this translation helpful? Give feedback.
-
Anyway, g24 has mtk bootloader They just skinned it. The code is all the same, but at least they made it look cooler. (I wonder if we could skin g23 one) I'll check tomorrow, but I'm sure I'll find the same vulnerable memcpy function as g23, that doesn't check for destination size. @R0rt1z2 (sorry for always pinging you), generally what would having a vulnerable memcpy function mean when dealing with phones? Do you think this can be used to cause a buffer overflow? 🤔 |
Beta Was this translation helpful? Give feedback.
-
I need someone to try all the functions in mtkclient: peek, poke, da efuse etc... If we have access to memwrite (I highly doubt it, but who knows), the phone would hey unlocked 100% because we could bypass secure boot measures for sure |
Beta Was this translation helpful? Give feedback.
-
It seems that from g24 and further, moto decided to give to this "Tinno" company the handling of spft, lamu (g15) has the same stuff and keys as g24 |
Beta Was this translation helpful? Give feedback.
-
@progzone122 100% confirmed that g24 does not have BROM: hard bricked g24 |
Beta Was this translation helpful? Give feedback.
-
While I was awaiting for the g23 script to be tried, I Decompiled the g24 lk binary.
Funnily enough, it doesn't even require a key to unlock the bootloader, there's just this check for secure boot.
Basically, if this check wasn't in place, it could be unlocked as easily as a Google pixel (apart from the 7 days connection it needs for not graying out oem unlock).
The only way to bypass this is making an exploit for disabling secure boot or try flashing seccfg through mtkclient.
We know g24 DA has some kind of access to seccfg thanks to the release notes present.
I'd need for someone to extract the spft from RSA first to confirm (I downloaded the one I used from the firmware archive) and also confirm this method to bypass DA SLA https://github.com/orgs/moto-penangf/discussions/16#discussioncomment-12017551
WHAT WE TRIED:
Beta Was this translation helpful? Give feedback.
All reactions